Compute Ontario
Compute Canada

 

HPCVL is a consortium of four universities led by Queen's University, and includes Carleton University, University of Ottawa, and the Royal Military College of Canada. We specialize in secure, advanced computing resources and support for academic and medical clients. HPCVL operates a high performance data centre as part of the Compute Canada family.

Highlights

Compute Canada Launches 2016 Research Platforms and Portals (RPP) Competition

Letters of Intent due by August 13, 2015

Researchers building platforms or portals to serve the needs of a community of users are invited to apply to Compute Canada’s 2016 Research Platforms and Portals (RPP) competition, which grants access to computing and storage resources on Compute Canada’s systems.

Don Aldridge to lead research computing laboratory

Don Aldridge, a former research executive at IBM Canada, has joined Queen’s University on a full-time basis as the executive director of the High Performance Computing Virtual Laboratory (HPCVL), and senior advisor, advanced computing and data analytics.

2015 Compute Ontario HPC Summer School

The Compute Ontario Summer School on High Performance and Technical Computing is an annual educational event for students, postdocs and researchers in the areas of computational science. Jointly organized by SHARCNET (Ontario-West), SciNet (Ontario-Central) and HPCVL (Ontario-East), attendees are provided with opportunities to learn and share knowledge and experience in high performance and technical computing.

This all-week series of workshops will cover a range of high-performance computing related subjects, such as Introductions to OpenMP, MPI, and CUDA programming of GPUs. The Summer School will be conducted in three instalments for the different regions of Ontario:

  • The West event is organized by SHARCNET and takes place May 25-29, 2015 at the Western University.
  • The Centre event is on July 13-17 at the the University of Toronto. It is organized by SciNet.
  • Finally, the East installment takes place at Queen's University, Kingston on July 27-31, and is organized by HPCVL.

 

HPCVL hosts the premiere Advanced Computing and Analytics in Medical Research Symposium, on May 12th, 2015 in Ottawa

HPCVL hosted the premiere "Advanced Computing and Analytics in Medical Research Symposium" on May 12th, 2015, which included several renowned speakers. Advances in technology have led to the availability of large volumes of data from a variety of research studies and clinical environments. This data can be processed, integrated, and stored to present opportunities for new understandings of disease and treatments, and has the potential to help save lives and billions of dollars throughout the health care system as well as potentially helping to create new products and methodologies. This symposium examined several aspects of this opportunity and the issues surrounding the use of this data. 

HPCS 2015 - Montreal - June 17-19th

HPCS 2015, Canada’s foremost supercomputing conference, will be co-hosted by Compute Canada and Calcul Québec at the SGW campus of Concordia University in downtown Montreal from June 17th to June 19th 2015. As usual, the three days of the main conference will be preceded by two days of workshops and meetings on June 15th and 16th.

International HPC Summer School 2015

Graduate students and postdoctoral scholars from institutions in Canada, Europe, Japan and the United States are invited to apply for the sixth International Summer School on HPC Challenges in Computational Sciences, to be held June 21-26, 2015, in Toronto, Canada. For more details, click here.

HPCVL Provides Secure Resources for Ontario Brain Institute (OBI)

HPCVL provides a secure environment and computing resources for the Ontario Brain Institute (OBI). The Ontario Brain Institute is creating an extensive brain research database called the Brain-CODE that will allow researchers to work faster and more efficiently. Read more here.

HPCVL Library for Parallel Programming on Multicore Clusters

The "Double-layer Master-Slave Model" (DMSM) combines an MPI-based approach for workload distribution on a cluster with local OpenMP-based task scheduling for multicore nodes. It is therefore ideally suited to exploit clusters with multicore and multithreaded nodes. At HPCVL, we have implemented this model in the form of a library that requires minimal user input. The library is freely available for download in the form of Fortran90 and C source code. Since only standard language features were used in the development, it can be deployed easily on multiple platforms. For details, see our FAQ file, or the User's Manual.